Prospective Data Show Comparable Performance Between AI and Dermatologists

Key Takeaways

• Prospective data show AI systems achieve diagnostic performance comparable to dermatologists in melanoma detection using dermoscopy, according to the results of a new analysis.
• AI-assisted dermatologists demonstrated higher sensitivity (91.9%) and specificity (83.7%) in the single available study.
• High risk of bias and limited generalizability remain key barriers to routine clinical adoption.

A new systematic review and meta-analysis of prospective studies suggests artificial intelligence (AI) systems can achieve diagnostic performance comparable to dermatologists in melanoma detection using dermoscopy.

Investigators evaluated the diagnostic performance of dermatologists, artificial intelligence (AI) systems, and AI-assisted clinicians in melanoma detection using dermoscopy for the analysis, which included 11 studies and more than 2,500 patients, along with 50 dermatologists. The authors, publishing in JAMA Dermatology, focused on prospective designs with histopathologic confirmation as the reference standard.

According to the analysis, dermatologists achieved a sensitivity of 78.6% (95% CI, 67.5% to 88.1%) and specificity of 75.2% (95% CI, 63.3% to 84.3%). AI systems demonstrated comparable performance, with a sensitivity of 80.9% (95% CI, 63.6% to 94.5%) and specificity of 75.6% (95% CI, 64.5% to 85.6%). In direct head-to-head comparisons, AI showed higher specificity with similar sensitivity relative to dermatologists. Just one prospective study evaluated AI-assisted dermatologists and reported improved outcomes with a sensitivity of 91.9% and specificity of 83.7%. 

Study quality was cited as a notable limitation. Most included studies were judged to have a high risk of bias, especially in patient selection and index test domains. Many studies relied on preselected lesions suspected of melanoma and binary classification frameworks, which may not be reflectibe of real-world clinical complexity.

“AI systems perform at comparable levels to dermatologists for melanoma diagnostics and may enhance performance when used as a decision-support tool,” the authors wrote. “However, the frequent risk of bias and limited generalizability of current studies highlight the need for broader validation in unselected patient populations in the clinical setting.”

Source: Laiouar-Pedari S, et al. JAMA Dermatology. Doi:10.1001/jamadermatol.2026.0217